ONLINE SUPPLEMENT Lubich et al.
Supplement Figure 1
1
Supplement Figure 1: General gating strategy
Presented is a representative gating strategy of the flow-cytometry-based technology to
analyze sub-visible particles in protein-based therapeutics. Size range gates were controlled
and if needed corrected using Fluoresbrite® YG Carboxylate Size Range Kit I & II [A].
CountBright™ Absolute Counting Beads were gated out using the 660/20 band-pass filter in
the red detector array [B]. Gating on protein sub-visible particles in the Bis-ANS channel
(violet detector array, 450/50 band-pass filter) was done by acquiring a non-protein subvisible particle control (silicon oil particles, preparation Non-Prot) as a negative control
reference [C] and protein sub-visible particle controls (biotinylated BSA-avidin complex,
preparation Prot and aggregated amyloid beta 1-40 peptide, preparation Prot-Crossß) as
positive controls [E and G]. Protein sub-visible particles bearing a cross-β-sheet structure
were analyzed in the DCVJ channel (blue detector array, 530/30 band-pass filter) using
preparation Non-Prot and preparation Prot as negative reference controls [D and F] for
cross-β-sheet structures and preparation Prot-Crossß as a positive control [H]. All size range
gates were applied for each staining gate.
2
Supplement Table 1A
amyloid-like protein
particles/ml
particles/ml
1
2
3
4
5
6
2,903,406
2,889,310
2,595,874
3,108,448
3,093,660
2,784,400
2,350,739
2,338,488
2,688,654
CV %
protein
particles/ml
run
10.32 %
2,666,654
CV %
1,765,834
10.36 %
2,495,028
3,101,138
3,086,665
2,963,407
2,700,258
2,690,203
2,486,555
CV %
16.32 %
Supplement Table 1B
run
particles/ml
CV %
1
2
3
4
5
6
2,999,795
2,862,464
2,939,847
3,072,418
2,740,746
2,855,702
3.92 %
protein
particles/ml
2,985,417
2,847,235
2,926,533
3,042,521
2,725,321
2,848,923
amyloid-like
protein particles/ml
2,665,343
2,607,556
2,608,210
2,710,293
2,570,223
2,575,184
CV %
3.79 %
CV %
2.05 %
Supplement Table 1C
0.75µm-1µm
run
particles/ml
1
2
3
4
5
6
892,144
1,023,858
704,284
852,891
1,038,624
855,401
1µm-2µm
CV %
particles/ml
13.87 %
263,112
309,896
208,040
277,445
309,597
235,712
2µm-4.5µm
CV %
particles/ml
15.18 %
160,163
168,474
124,717
213,051
214,182
164,945
4.5µm-6µm
CV %
particles/ml
19.67 %
20,257
20,603
14,227
24,287
26,347
19,461
6µm-10µm
CV %
particles/ml
CV %
20.13 %
10,314
11,359
7,938
11,561
14,780
10,014
20.56 %
Supplement Tables 1A-1C: Precision and accuracy of the flow-cytometry based technology
The precision of the method was evaluated by assessing the inter- and intra-assay variations.
Aggregated amyloid beta 1-40 peptide (preparation Prot-Crossß) was used as a positive
control for analyzing concentrations (in particles/ml) of total sub-visible particles, protein
sub-visible particles and cross-β-sheet protein particles. We analyzed preparation ProtCrossß in 6 independent experiments for inter-assay variation [Table 1A] and in 6 parallel
experiments on the same day for intra-assay variation [Table 1B]. Each sample was
measured in 3 replicates in each experiment and mean numbers of sub-visible particles
contained in 1 ml were calculated taking the dilution factor into account. This approach was
used to calculate numbers of total sub-visible particles, protein and protein containing sub3
visible particles and cross-β-sheet protein particles. All mean values for the inter-assay and
for the intra-assay were used to calculate the coefficient of variation (CV %) for each
property. We observed that the CV % of all the properties tested ranged between 10.32 %
and 16.32 % for the inter-assay variations and between 2.05 % and 3.92 % for the intra-assay
variations. In addition, we tested the accuracy of the method regarding the size range gates
by calculating the CV % of total sub-visible particle counts per ml for each size gate [Table
1C]. We met the acceptance criteria that were set to a maximum of 25 % for each CV which
complies with regulatory guidelines [38, 39].
4